I. Who Is Listening to the Children’s Voices?
An eight-year-old girl’s parents are divorced and fighting over custody. During the week she lived with her father, she wrote a letter saying, “I want to live with Dad.” During the week she lived with her mother, she recorded a video saying, “I want to live with Mom.” Both pieces of “evidence” were laid out on the judge’s desk—each unequivocal, yet each contradicting the other.
The judge pored over these materials repeatedly, yet found it difficult to determine which one truly reflected the child’s genuine thoughts. This is because these “expressions of will” were manufactured—every moment spent alone with one parent was an opportunity to express a stance; behind every letter and every video stood an adult desperate to win custody. The child’s true voice, in fact, was drowned out by the noise created by both parents.
This is not an isolated case. Parents certainly love their children, but loving a child and representing the child in a family law dispute are two different things. Love would not lead parents to intentionally harm their child, but the stance taken in litigation can cause parents to unconsciously embed the child’s interests within their own demands—this is not a moral issue, but a matter of perspective. Regardless of which party represents the child in the proceedings, the child’s interests will inevitably be “embedded” within that party’s own demands. When both sides are “speaking on behalf of the child,” the child’s own interests paradoxically become the first to be sacrificed.
Children need an independent voice that belongs to neither parent. Yet under the current institutional framework, there is no formal channel for this voice to be heard.
In fact, this dilemma is not unique to China. When the interests of parents conflict with those of their children, the question of who should represent the child has been answered differently across various legal jurisdictions.
II. Lessons from Other Jurisdictions: Three Institutional Logics
Looking at practices across various jurisdictions, the design logic of child interest representation systems can generally be divided into three models: the independent representative model, the legal representative model, and the investigative support model. The core difference among these three models lies in the degree of the representative’s independence and the presence or absence of legal representation authority. There is one key distinction: Does the representative speak on behalf of the child, or does the representative litigate on the child’s behalf? In the former model, the representative operates independently of the adversarial litigation process and directly articulates the child’s interests to the court; in the latter, the representative is embedded within the litigation process and protects the child’s rights through legal representation. This distinction determines the direction of institutional design.
Independent Representative Model: Taiwan, Singapore
Taiwan—The System of Procedural Supervisors
Taiwan’s *Family Affairs Act*, which took effect in 2012, drew upon Germany’s system of procedural assistants (Verfahrensbeistand) to establish the system of procedural supervisors, which is currently the most mature system for representing the best interests of children in the Chinese-speaking legal sphere. According to research by Song Hanlin, this system is based on the principles of state parental authority and the maximization of the interests of minors [1].
Procedural guardians are appointed by the court upon application or ex officio in situations where there is a conflict of interest between the minor and their legal representative, or where the minor is unable to independently exercise their procedural rights. Regarding the selection of candidates, the legislative rationale for Article 16 of the Family Matters Act states that the appointment of a procedural guardian involves a significant degree of public interest and professionalism; therefore, the position should be filled by personnel from social welfare authorities or social welfare institutions, or by individuals recommended by bar associations or social worker associations who possess professional knowledge relevant to handling family matters [2]. Candidates must possess an awareness of gender equality, respect for cultural diversity, and experience in handling family matters.
In terms of authority, as a legal entity with independent legal status, the procedural guardian enjoys the rights necessary to maximize the interests of the minor: appearing in court on behalf of the minor child, presenting arguments, conducting investigations, performing procedural acts on the child’s behalf, and participating in mediation proceedings [1]. Crucially, the procedural guardian is independent of both parents—he or she is not the agent of either party, but rather the advocate for the child’s own interests.
According to Zhang Run’s systematic study on the practical effectiveness of this system, the procedural guardian system has established a relatively stable operational mechanism in practice [3].
Singapore—The Child Representative System
Part 4, Division 1 of Singapore’s Family Justice Rules 2014 specifically provides for the Child Representative system, under which the court appoints a representative to act in the best interests of the child in legal proceedings involving the child’s custody and welfare.
A unique feature of the Singaporean system is that the Child Representative does not operate in isolation but is embedded within the framework of therapeutic jurisprudence—a judicial philosophy that seeks to repair relationships rather than determine a winner or loser. According to published academic literature, Singapore is the first country in the world to explicitly and comprehensively apply therapeutic jurisprudence in the adjudication of family disputes. It released the “Model for Therapeutic Jurisprudence in Family Justice Courts” in October 2024, which was revised in March 2026 [4]. This means that the role of the Child Representative is not to “win the case for the child” in adversarial litigation, but rather to promote solutions that are in the child’s best interests within a restorative framework. In addition, the role of Parenting Coordinator—introduced in the 2018 amendment—complements that of the Child Representative: the former mediates parenting disputes between parents, while the latter represents the child’s interests; both fulfill their respective roles within the restorative justice framework.
Taiwan and Singapore share the common feature that the representative is independent of both parents and presents opinions directly to the court; the core function is not legal representation but the independent articulation of the child’s interests. The difference lies in the fact that Taiwan’s Procedural Supervisor places greater emphasis on independent advocacy within the litigation process, whereas Singapore’s Child Representative is embedded within a broader restorative justice system.
Legal Representation Model: Hong Kong, the United States, and Australia
Under this model, the representative must be a licensed attorney and protect the child’s rights through legal representation. However, the degree of independence varies across the three jurisdictions: Hong Kong’s Independent Legal Representative and the U.S. GAL (Guardian ad Litem) lean more toward the traditional role of legal representation, while Australia’s ICL (Independent Child Representative)—though also required to be held by an attorney—is closer in terms of independence and institutional design to the independent representative model; it can be regarded as the variant with the highest degree of independence within the legal representation model.
In June 2023, the Hong Kong Legislative Council enacted the Family Proceedings Ordinance, which provides comprehensive provisions on family court jurisdiction and legal procedures and established the system of Independent Legal Representatives for Children [5]. Unlike Taiwan, Hong Kong’s Independent Legal Representatives are lawyers who are licensed to practice law—essentially providing legal representation for the child rather than independently expressing the child’s interests. At the same time, Hong Kong has established detailed guidelines for judges’ meetings with children to ensure that judges can hear the children’s views directly, which to some extent addresses the shortcoming of the legal representation model where children are “represented” rather than “heard.”
United States—The Guardian ad Litem System
The Guardian ad Litem (GAL) system in the United States exhibits greater internal variation. A GAL is appointed as the guardian and legal representative of a minor child and participates in litigation on behalf of the child’s interests [6]. However, practices vary by state: some states require that a GAL be an attorney, while others allow social workers or professionals with a background in psychology to serve in this role. Family courts in places such as Travis County have dedicated GAL teams. The authority of a GAL is quite broad—they participate in litigation as a party to the case, receive copies of all pleadings and documents, are notified prior to any action affecting the minor child, file pleadings and reports, participate in agreements or settlements involving the child’s best interests, and inspect and copy records related to the child and parents (including hospital, school, and psychological records) [6].
It is worth noting that GALs fulfill a dual role: they are both the child’s litigation guardian and an officer of the court [7]. This dual identity carries the risk of a conflict of interest—is the GAL speaking on behalf of the child, or conducting an investigation on behalf of the court?
Even more noteworthy are the findings of empirical research. Some studies indicate that, overall, the existence of GALs has not produced beneficial effects that meet research standards [8]. This conclusion serves as a wake-up call: having a system does not equate to the system being effective. The design of the system is more important than its mere existence—if the representative’s role is ambiguous and the boundaries of their authority and responsibilities are unclear, even an established system may fail to truly protect the best interests of the child.
Australia—The Independent Children’s Lawyer System
Australia’s Independent Children’s Lawyer (ICL) system has evolved within the family court system established under the Family Law Act 1975 and is one of the longest-running independent child representation systems in the world. The legal basis for the system is Section 68L of the Family Law Act, and ICLs are appointed by the court by order [9].
One core design feature of the ICL system deserves special attention: the ICL represents the child’s “best interests,” not the child’s “wishes.” This means that if the ICL believes the wishes expressed by the child are not in the child’s best interests, the ICL may take a position that differs from the child’s wishes—but must present both the child’s wishes and the ICL’s assessment of the best interests to the court [9]. This distinction cuts to the heart of the dilemma described at the outset: the conflicting “wishes of the child” presented to the judge demonstrate precisely that “wishes” can be manufactured, whereas “best interests” require independent professional judgment.
The 2023 Family Law Amendment Act made significant changes to the ICL system: it introduced a legislative requirement that ICLs must meet with children directly—previously, whether an ICL met with a child was left to the discretion of the lawyer, but following the amendment, it became a statutory obligation [10]. This amendment demonstrates that even a system that has been in operation for nearly fifty years requires continuous refinement in practice—a system is not a one-time design but rather a process of continuous iteration.
In terms of funding, the work of ICLs is funded by the Legal Aid Commissions of Australia’s states and territories [9], with the National Legal Aid providing unified support for resources, training, and professional development. This arrangement directly addresses the “lack of funding guarantees” issue faced by domestic pilot programs—the legal aid system can serve as the financial backbone of the child’s interests representative system.
The Japan Federation of Bar Associations sent a fact-finding mission to Australia precisely because the ICL system is the most mature within the common law tradition and complements the “procedural guardian” system in Taiwan, which is based on the civil law tradition—Japan was seeking reference points from different legal systems to inform its institutional design.
Investigation Support Model: Japan
Japan has not yet established a formal system of child representatives, and the Japan Federation of Bar Associations (JFBA) is currently promoting legislative discussions on the matter [9]. Under the existing system, investigators at family courts (family court investigators) perform some functions related to investigating the best interests of the child—they investigate the facts of family cases and provide investigative reports to judges—but they do not possess independent standing in litigation to represent the child’s interests [6].
In other words, Japanese family court investigators serve as investigative assistants to the court, not as advocates for the child. They provide information to the judge but do not speak on behalf of the child in litigation. The reform being promoted by the JFBA aims to shift from this role of “investigative assistant” to that of “independent representative”—the proposed children’s advocate would possess legal expertise and participate independently in proceedings to represent the interests of minors [9]. The Japan Federation of Bar Associations (JFBA) previously dispatched a fact-finding mission to Australia to study the Independent Children’s Lawyer (ICL) system, with the aim of gathering experience to advance legislative efforts [10].
The situation in Japan bears a certain resemblance to that in mainland China: both follow the civil law system, both face the challenge of transitioning from investigative assistance to independent representation, and both are currently in the “work in progress” stage. The Japan Federation of Bar Associations’ dispatch of a fact-finding mission to Australia to study the ICL system was precisely an effort to seek references from different legal systems for institutional design—Taiwan’s procedural supervisors provide a mature reference within the civil law system, while Australia’s ICLs provide a mature reference within the common law system, and the two are complementary.
A Comparison of the Three Models
Looking back at the three models, the core distinction lies in the answer to one question: Whom does the representative actually represent?
Under the independent representative model, the representative acts in the child’s own best interests, independent of both parents, and presents arguments directly to the court. Under the legal representative model, the representative protects the child’s rights through legal representation; however, the representative’s role is more easily dominated by “legal technicalities,” and the child’s own voice may be overshadowed by the representative’s arguments. Under the investigative assistant model, the representative is not even the child’s spokesperson but rather an investigative assistant to the judge—the protection of the child’s interests depends on the judge’s judgment rather than on an independent representative.
This comparison offers an important insight: the key to institutional design lies not in whether the role of “representative” is established, but in whether that role is truly independent of both parents and capable of directly expressing the child’s interests. Empirical research on U.S. GALs (Guardians ad Litem) confirms precisely this point—a “representative” with an ambiguous role and unclear boundaries of authority and responsibility may be worse than having no representative at all.
III. Our Path
Existing Initiatives
This issue has not gone unnoticed in China. In November 2017, the Shanghai High People’s Court convened an expert symposium on “The Theory and Practice of Child Rights Representatives,” taking the lead in exploring the “child rights representative” system. Applicable scenarios include cases where parents have abandoned or abused their minor children, where parents insist on relinquishing custody of their children, and where the property rights of minor children may be infringed upon. Candidates are primarily drawn from officials of the district Women and Children’s Working Committees, and their responsibilities include preliminary communication and investigation, participating in mediation and court hearings, expressing opinions to protect the rights and interests of minors, and continuously monitoring the minors’ physical and mental well-being and healthy development.
Social workers from the Shanghai Sunshine Community Youth Affairs Center also participate in pre-trial mediation of family cases, family investigations, psychological counseling, follow-up visits and supervision, family education guidance, serving as rights representatives, and acting as visitation supervisors [11]. The role of social workers as “child advocates” is defined as follows: adopting an objective “third-person” perspective to conduct field investigations into actual circumstances, listen to children’s inner thoughts, and give voice to their genuine needs [12].
In July 2024, the Shanghai High People’s Court established a Comprehensive Trial Division for Minors and Family Cases, creating pioneering judicial mechanisms—such as social investigations and the appointment of child rights representatives—that are the first of their kind in the country [13]. In May 2026, the Pudong New Area Court and the District Bureau of Justice jointly launched the “Spring Dandelion: Legal Aid to Protect and Nurture Children’s Growth” direct legal aid support mechanism for juvenile trials, promoting a shift in legal aid from “acceptance upon application” to “proactive service” and establishing a full-chain service system encompassing “case representation + psychological rehabilitation + family healing” [14]. Although this mechanism is still in its early stages and its effectiveness remains to be seen, the concept of “proactive service” is already worthy of attention.
What’s Still Missing
Efforts are underway, but there remains a significant gap before this becomes institutionalized. The recommendations from the Shanghai CPPCC and the proposal by Liu Yan, a member of the National Committee of the Chinese People’s Political Consultative Conference (CPPCC), both point to the same pain points [15][16]:
1. Unclear rights and responsibilities. What is the legal status of a children’s rights representative in litigation? Are they a party to the case, a legal representative, or an independent third party? To what extent are the representative’s opinions binding on the judge? There are currently no clear legal provisions addressing these issues, leaving representatives in an awkward position during court proceedings—if they speak, their words may carry no weight; if they remain silent, they are derelict in their duties.
2. Lack of funding guarantees. The work of these representatives entails costs—investigations, site visits, court appearances, and follow-up efforts all require time and professional expertise. However, there is currently no institutionalized source of funding; pilot programs often rely on project grants or volunteer efforts, which is unsustainable.
3. Lack of Professionalism. Current appointees primarily come from cadres of the Women and Children’s Work Committees and social workers; while they have social work experience, they may not possess legal expertise or the ability to participate in litigation. In contrast, Taiwan’s “procedural supervisors” are selected from candidates recommended by the Bar Association and the Association of Social Workers, and are required to have experience handling family law cases—this difference in professional standards directly impacts the quality of the system’s operation.
4. A strong emphasis on ad hoc handling. Pilot programs in Shanghai, Nanjing, and other locations still carry a strong “case-by-case” character and have not been implemented on a routine basis [16]. If a system cannot transition from “ad hoc handling” to “routine application,” it will be difficult to truly transform the ecosystem.
Directions for Institutional Development
Drawing on experiences from multiple jurisdictions and taking domestic realities into account, the development of this institutional framework must address three core questions:
First, what does it represent? — It represents the “best interests,” not the “wishes.”
This is the most valuable aspect of Australia’s ICL system: ICL represents the child’s “best interests,” not the child’s “wishes.” The two may sometimes align, and sometimes conflict. An eight-year-old child saying, “I want to live with my mom,” expresses a wish; however, if the mother suffers from a serious mental illness or the home environment is detrimental to the child’s development, the child’s best interests may run counter to that wish.
Child representation systems in all jurisdictions are based on the fundamental principle of “maximizing the child’s best interests,” as required by the United Nations Convention on the Rights of the Child. However, “maximizing best interests” is a value-driven guideline; at the level of institutional design, we must further ask: What exactly does the representative represent? If the representative represents “wishes,” then the dilemma in the opening scenario cannot be resolved—a child’s wishes can be fabricated, guided, or manipulated by parents, and the representative’s mere repetition of those wishes amounts to nothing more than affixing a professional seal to yet another piece of “evidence.” Only when the representative represents the “best interests” can the dilemma of manipulated wishes be overcome—the representative’s duty is not to repeat what the child has said, but to present to the court, based on independent professional judgment, what is best for the child.
Of course, this does not mean that the child’s wishes are unimportant. Under Australia’s ICL system, if the child’s wishes differ from the ICL’s assessment of the child’s best interests, the ICL must present both to the court simultaneously, leaving it to the judge to make a comprehensive determination. The ingenuity of this design lies in the fact that it both acknowledges the child’s right to express their views and prevents “wishes” from becoming the sole basis for the court’s decision.
The independent representative model (Taiwan, Singapore) better ensures the independent expression of a child’s interests than the legal representative model (Hong Kong, the United States). In Chinese family court proceedings, judges already bear the responsibility of conducting ex officio investigations; if the representative were merely a legal representative, there would be a risk of overlap with the functions of an attorney. However, if the representative were an independent third party who directly presents the child’s best interests to the court, this would fill the gap in the current system where “the child has no voice of their own.” Taiwan’s procedural supervisor system, drawing on the German model, over a decade of experience in the Chinese legal system, and nearly five decades of practice and continuous refinement of Australia’s ICL system, provides a solid reference for China’s institutional design.
Second, who should serve as the representative? — Primarily professional attorneys, supplemented by a diverse range of professionals.
Regarding the selection of candidates, Taiwan’s system of procedural guardians offers a diverse range of options—attorneys, social workers, and social welfare professionals may all be recommended for appointment. However, Australia’s ICL system provides a different answer: the representative must be an attorney. Both approaches have their merits, but given the actual circumstances of family court proceedings in China, having a professional attorney serve as the representative for the child’s best interests is more feasible.
Having a lawyer represent the child may sound like adding yet another lawyer to the litigation process—how is this different from the lawyers hired by each parent? The distinction lies in the fact that this lawyer does not represent the claims of either party but solely represents the child’s best interests; the lawyer is not helping the child “win” the case but is helping the judge “see the situation clearly.”
There are three reasons for this.
First, family litigation is, by its very nature, a legal proceeding. The representative must present arguments, participate in the examination of evidence, and file motions during court hearings—procedural actions that require legal expertise to carry out effectively. While social workers and psychologists can play an irreplaceable role in areas such as investigation and counseling, legal expertise is the foundation for a representative to fulfill their duties within the litigation process.
Second, lawyers are bound by professional standards and ethics, with clearer boundaries between their powers and responsibilities and more explicit accountability mechanisms—which is precisely what is most needed to address the current pilot program’s pain point of “unclear powers and responsibilities.”
Third, the operational experience of Australia’s ICL system demonstrates that having a lawyer serve as the representative does not prevent the system from addressing the child’s psychological needs and emotional state—the 2023 legislative amendment requiring ICLs to meet directly with children precisely builds upon legal expertise by incorporating respect for the child’s agency.
Of course, having a lawyer serve as the representative does not mean excluding other professionals. On the contrary, the representative’s work requires the cooperation of professionals such as social workers and psychologists—tasks such as observing children during investigations and home visits, conducting psychological assessments, and evaluating the family environment all necessitate interdisciplinary collaboration. However, the role of “representative” in litigation proceedings should be assumed by a lawyer.
Third, how can this be ensured? — Legislative support is the foundation, and securing funding is essential.
The proposal put forward by the Shanghai CPPCC hits the nail on the head: clearly define the legal standing of minors as parties to litigation to provide legislative support for the system; adopt an open approach to bringing in professional expertise, build a talent pipeline, and secure funding sources; and clarify the powers and responsibilities of various departments to collaborate in establishing a comprehensive protection mechanism for minors [15]. Among these, legislative support is fundamental—without a legal basis, the role of children’s representatives will forever remain in the “pilot” and “exploratory” stages and cannot become a standard practice in family court proceedings.
Funding, however, is the key to whether the system can actually function effectively. Currently, funding for pilot programs relies either on project grants or on volunteer efforts, neither of which is sustainable. Australia’s approach offers a transferable model: the work of Independent Children’s Lawyers (ICLs) is funded by the legal aid commissions of each state and territory, with the National Legal Aid providing unified resources, training, and professional development support. The legal aid system itself already has institutionalized funding sources and professional management mechanisms; incorporating funding for Children’s Interest Representatives into the legal aid system is more stable than project funding and more reliable than volunteer efforts—this is a path that can be adopted.
Japan and China face similar situations: both are transitioning from “investigative assistance” to “independent representation.” However, Japan benefits from the systematic promotion by the Japan Federation of Bar Associations and legislative preparations following a study tour to Australia; China has accumulated nearly a decade of pilot experience in places like Shanghai, along with recommendations and proposals at the levels of the Chinese People’s Political Consultative Conference and the National People’s Congress. These fragmented efforts require an institutionalized outlet.
In family disputes, a child’s interests should not be a bargaining chip in parental conflicts, nor should they be conflicting “evidence” on a judge’s bench. Experience across multiple jurisdictions has proven that the system of child representatives is not a luxury but a necessary component of family justice—the question is not “whether to establish it,” but “how to establish it so that it truly works.”
参考文献
[1] 宋汉林：《台湾程序监理人制度述评及其启示——以未成年人利益最大化为中心》，载《青少年犯罪问题》2014年第5期。
[2] 姜世明：《台湾家事事件审理制度之立法特色与法理争议》，收于许少波主编《法治的布道者》。
[3] 张润：《台湾程序监理人制度：规范构造、运行实效及其启示》，载《法律科学》2017年第2期。
[4] 新加坡家事司法法院：《家事司法法院愈疗性司法模型》（Family Justice Courts Therapeutic Justice Model），2024年10月发布，2026年3月修订。
[5] 香港《家事诉讼程序条例》，2023年6月立法会制定发布。
[6] 参见万方学位论文：《论日本家事司法中未成年人程序代理人制度及其对我国的启示》。
[7] 美国GAL作为法院官员（officer of the court）的双重角色，参见相关州法院任命令规范。
[8] 有实证研究指出GAL制度在整体上未产生符合研究标准的有益效果，具体研究出处待进一步查证。
[9] 澳大利亚《家庭法1975》（Family Law Act 1975， Cth）第68L条；ICL指南参见National Legal Aid发布的相关规范。
[10] 澳大利亚2023年《家庭法修正案》（Family Law Amendment Act 2023）。
[11] 上海市阳光社区青少年事务中心社工参与家事案件工作，参见相关报道。
[12] 社工担任“儿童代言人”的角色定位，参见相关报道。
[13] 上海高院2024年7月设立未成年人与家事案件综合审判庭，参见相关报道。
[14] 浦东新区法院“春天的蒲公英·法援护苗成长”未成年人审判法律援助直通护航机制，2026年5月启动。
[15] 上海政协关于建立和完善未成年人权益代表人制度的建议。
[16] 在沪全国政协委员刘艳关于建立和完善未成年人权益代表人制度的建议。
